The market for wireless devices has been in continual expansion as communication network architecture and services have developed and evolved over recent years. Uptake of Machine Type Communication (MTC) devices in particular is expected to grow rapidly in the coming years. MTC devices are wireless devices generally associated with equipment as opposed to a human user. Such devices typically include sensors, actuators or meters which are connected to the internet or a computer cloud via a communication network. For example, increasing numbers of residential and commercial buildings are being equipped with “Smart” electric, gas and water meters that are ale to periodically report energy, gas or water consumption without the need for a site visit from a technician.
In many cases, MTC devices will have a fixed installation location and will be battery powered, to allow for easier and more flexible installation. The desired operation time for a single battery or battery charge is typically several years, to avoid costly or inconvenient battery replacements or charging. Typical deployment scenarios for MTC devices are illustrated in FIGS. 1 and 2. Referring to FIG. 1, in a first example, an MTC device 2 may communicate directly with a local access point (AP) 4, or radio base station. For simplicity, both the local access point and radio base station are referred to as AP in the following discussion, and the deployment is known as a device-to-AP (D2AP) installation. Referring to FIG. 2, in another example, the MTC device 2 may communicate directly with another device 6, which may be a Smartphone or another MTC device. The second device 6, known as a master device, may communicate with another network node such as an AP 4. This deployment is known as a device-to-device (D2D) installation.
In either of the illustrated installations, communication network traffic is often characterised by small amounts of data, of the order of a few bytes, transmitted or received relatively infrequently. In order to compete with other short range radio transceivers, and to preserve battery life, MTC devices are generally designed to be simple and low cost, stripped of costly power consuming auxiliary features found on more advanced wireless devices, and providing only the minimum functionality to enable transmission of the relevant information to the network.
Guidance for MTC device installation is generally limited to basic rules of thumb covering location and direction of antennas for best signal reception. Such rules of thumb may be that the antenna should be pointing upward, and the device should be close to a window. This guidance is relatively basic, particularly considering that for devices having a fixed installation, installation location may affect the power required for radio communication during the lifetime of the device, and hence may impact the battery life of the device.
An additional difficulty surrounding battery lifetime for wireless devices, and in particular MTC devices, is the need to manage battery replacement. For devices which may have a battery lifetime extending several years, it is helpful to inform a suitable person, for example a customer or user of the device, or a network or service provider, when it is nearing time to replace the battery, in order to avoid unnecessary down time. Remaining battery life in cellular communication devices like Smartphones is typically computed solely based on average power consumption since the start of the device or since its last charge. The average power consumption is estimated, and used together with the current charge level to estimate the remaining battery lifetime. A problem with predicting future battery life based on power consumed so far in a device is that a body of historical usage data is required to make any kind of prediction as to future usage. At installation time, when the programming of battery replacement might be contemplated, insufficient data will be available for accurate prediction. For example, if a gas meter reports gas consumption once per day, several days of communication would be needed before even a rough estimate of the power consumption, and hence battery, could be predicted.